The present invention relates generally to computers, and more particularly, to computer storage systems using data replication for high reliability systems.
A parallel sysplex or a system complex is a cluster of mainframe computers acting together as a single system image with an operating system such as z/OS®. A parallel sysplex is an example of a high-reliability system that allows multiple logical partitions to communicate and co-ordinate synchronized data storage and access for large-scale data storage. A parallel sysplex provides data sharing capabilities for accessing multiple databases to read and write as shared data. In applications, such as financial transactions requiring high-reliability systems, it is important to have multiple copies of data. High-reliability systems that utilize high-reliability storage systems typically use data replication to maintain a secondary copy (e.g., a secondary volume stored in one or more secondary devices) of the data stored in a primary volume on one or more primary devices. Computer systems or sysplex members requiring high-reliability storage systems typically employ data replication technologies, such as data mirroring, disk mirroring, data shadowing, or other similar replication schemes. Data mirroring technologies, such as peer-to-peer remote copy (PPRC) are used to keep data synchronized between at least two devices. Operating systems, such as z/OS and other operating systems employ data mirroring techniques that are controlled by replication or mirroring management software to improve data availability and to prevent data loss.
In some cases, planned or unplanned swapping between primary and secondary volumes occurs. A swapping function allowing the designation of the secondary volume as the primary volume is supported in high-reliability storage systems using various software functions and performs the swap in all members of the system complex. In this case, a secondary volume stored in one or more secondary devices becomes the new primary volume. In the event of a system or storage device failure, recovery can be initiated automatically with minimal or no data loss.